CN112820397B - Method for establishing peri-operative risk prediction model of coronary artery bypass grafting - Google Patents
Method for establishing peri-operative risk prediction model of coronary artery bypass grafting Download PDFInfo
- Publication number
- CN112820397B CN112820397B CN202110077755.1A CN202110077755A CN112820397B CN 112820397 B CN112820397 B CN 112820397B CN 202110077755 A CN202110077755 A CN 202110077755A CN 112820397 B CN112820397 B CN 112820397B
- Authority
- CN
- China
- Prior art keywords
- risk factors
- risk
- prediction
- surgery
- model
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Active
Links
- 238000000034 method Methods 0.000 title claims abstract description 26
- 210000004351 coronary vessel Anatomy 0.000 title claims abstract description 21
- 238000013058 risk prediction model Methods 0.000 title claims abstract description 12
- 206010019280 Heart failures Diseases 0.000 claims abstract description 16
- 238000001356 surgical procedure Methods 0.000 claims description 51
- DDRJAANPRJIHGJ-UHFFFAOYSA-N creatinine Chemical compound CN1CC(=O)NC1=N DDRJAANPRJIHGJ-UHFFFAOYSA-N 0.000 claims description 40
- 102100036475 Alanine aminotransferase 1 Human genes 0.000 claims description 28
- 108010082126 Alanine transaminase Proteins 0.000 claims description 28
- 101800000407 Brain natriuretic peptide 32 Proteins 0.000 claims description 25
- 102400000667 Brain natriuretic peptide 32 Human genes 0.000 claims description 25
- 101800002247 Brain natriuretic peptide 45 Proteins 0.000 claims description 25
- HPNRHPKXQZSDFX-OAQDCNSJSA-N nesiritide Chemical compound C([C@H]1C(=O)NCC(=O)N[C@@H](CCCNC(N)=N)C(=O)N[C@@H](CCCCN)C(=O)N[C@@H](CCSC)C(=O)N[C@@H](CC(O)=O)C(=O)N[C@@H](CCCNC(N)=N)C(=O)N[C@H](C(N[C@@H](CO)C(=O)N[C@@H](CO)C(=O)N[C@@H](CO)C(=O)N[C@@H](CO)C(=O)NCC(=O)N[C@@H](CC(C)C)C(=O)NCC(=O)N[C@@H](CSSC[C@@H](C(=O)N1)NC(=O)CNC(=O)[C@H](CO)NC(=O)CNC(=O)[C@H](CCC(N)=O)NC(=O)[C@@H](NC(=O)[C@H](CCSC)NC(=O)[C@H](CCCCN)NC(=O)[C@H]1N(CCC1)C(=O)[C@@H](N)CO)C(C)C)C(=O)N[C@@H](CCCCN)C(=O)N[C@@H](C(C)C)C(=O)N[C@@H](CC(C)C)C(=O)N[C@@H](CCCNC(N)=N)C(=O)N[C@@H](CCCNC(N)=N)C(=O)N[C@@H](CC=1N=CNC=1)C(O)=O)=O)[C@@H](C)CC)C1=CC=CC=C1 HPNRHPKXQZSDFX-OAQDCNSJSA-N 0.000 claims description 25
- 229940109239 creatinine Drugs 0.000 claims description 20
- 230000002861 ventricular Effects 0.000 claims description 19
- 230000034994 death Effects 0.000 claims description 17
- 231100000517 death Toxicity 0.000 claims description 17
- 230000006016 thyroid dysfunction Effects 0.000 claims description 17
- 230000004217 heart function Effects 0.000 claims description 16
- 238000007675 cardiac surgery Methods 0.000 claims description 15
- 239000008280 blood Substances 0.000 claims description 11
- 210000004369 blood Anatomy 0.000 claims description 11
- 102000001554 Hemoglobins Human genes 0.000 claims description 9
- 108010054147 Hemoglobins Proteins 0.000 claims description 9
- 238000004458 analytical method Methods 0.000 claims description 9
- 208000031226 Hyperlipidaemia Diseases 0.000 claims description 8
- 230000007213 cerebrovascular event Effects 0.000 claims description 8
- 206010012601 diabetes mellitus Diseases 0.000 claims description 8
- 208000010125 myocardial infarction Diseases 0.000 claims description 8
- 238000012216 screening Methods 0.000 claims description 7
- 238000007477 logistic regression Methods 0.000 claims description 6
- 206010003658 Atrial Fibrillation Diseases 0.000 claims description 5
- 230000000747 cardiac effect Effects 0.000 claims description 5
- 238000002513 implantation Methods 0.000 claims description 5
- 230000003902 lesion Effects 0.000 claims description 5
- 230000002093 peripheral effect Effects 0.000 claims description 5
- 238000000611 regression analysis Methods 0.000 claims description 5
- 230000000391 smoking effect Effects 0.000 claims description 5
- 206010020772 Hypertension Diseases 0.000 claims description 4
- 238000011160 research Methods 0.000 claims description 4
- 230000004083 survival effect Effects 0.000 claims description 4
- 206010003662 Atrial flutter Diseases 0.000 claims description 3
- 238000011119 multifactor regression analysis Methods 0.000 claims description 3
- 230000002596 correlated effect Effects 0.000 claims description 2
- 238000010219 correlation analysis Methods 0.000 claims description 2
- 201000010099 disease Diseases 0.000 claims description 2
- 208000037265 diseases, disorders, signs and symptoms Diseases 0.000 claims description 2
- 238000000556 factor analysis Methods 0.000 abstract description 3
- 230000008569 process Effects 0.000 abstract description 2
- 238000012795 verification Methods 0.000 description 10
- 210000001685 thyroid gland Anatomy 0.000 description 6
- 208000029078 coronary artery disease Diseases 0.000 description 3
- 230000000694 effects Effects 0.000 description 3
- 238000002474 experimental method Methods 0.000 description 3
- 230000006870 function Effects 0.000 description 3
- 108010023302 HDL Cholesterol Proteins 0.000 description 2
- 108010028554 LDL Cholesterol Proteins 0.000 description 2
- 230000001746 atrial effect Effects 0.000 description 2
- 230000003416 augmentation Effects 0.000 description 2
- 210000004204 blood vessel Anatomy 0.000 description 2
- HVYWMOMLDIMFJA-DPAQBDIFSA-N cholesterol Chemical compound C1C=C2C[C@@H](O)CC[C@]2(C)[C@@H]2[C@@H]1[C@@H]1CC[C@H]([C@H](C)CCCC(C)C)[C@@]1(C)CC2 HVYWMOMLDIMFJA-DPAQBDIFSA-N 0.000 description 2
- 230000000875 corresponding effect Effects 0.000 description 2
- 230000006589 gland dysfunction Effects 0.000 description 2
- 230000006872 improvement Effects 0.000 description 2
- UFTFJSFQGQCHQW-UHFFFAOYSA-N triformin Chemical compound O=COCC(OC=O)COC=O UFTFJSFQGQCHQW-UHFFFAOYSA-N 0.000 description 2
- 206010002383 Angina Pectoris Diseases 0.000 description 1
- 206010010071 Coma Diseases 0.000 description 1
- 208000032928 Dyslipidaemia Diseases 0.000 description 1
- 208000017170 Lipid metabolism disease Diseases 0.000 description 1
- 208000005736 Nervous System Malformations Diseases 0.000 description 1
- 206010042434 Sudden death Diseases 0.000 description 1
- 230000005856 abnormality Effects 0.000 description 1
- 210000003169 central nervous system Anatomy 0.000 description 1
- 235000012000 cholesterol Nutrition 0.000 description 1
- 238000010586 diagram Methods 0.000 description 1
- 230000035487 diastolic blood pressure Effects 0.000 description 1
- 238000011156 evaluation Methods 0.000 description 1
- 230000007774 longterm Effects 0.000 description 1
- 206010025482 malaise Diseases 0.000 description 1
- 230000007246 mechanism Effects 0.000 description 1
- 239000000203 mixture Substances 0.000 description 1
- 210000003516 pericardium Anatomy 0.000 description 1
- 238000007619 statistical method Methods 0.000 description 1
- 230000035488 systolic blood pressure Effects 0.000 description 1
- 238000010200 validation analysis Methods 0.000 description 1
- 208000003663 ventricular fibrillation Diseases 0.000 description 1
- 206010047302 ventricular tachycardia Diseases 0.000 description 1
Classifications
-
- G—PHYSICS
- G16—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR SPECIFIC APPLICATION FIELDS
- G16H—HEALTHCARE INFORMATICS, i.e. INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR THE HANDLING OR PROCESSING OF MEDICAL OR HEALTHCARE DATA
- G16H50/00—ICT specially adapted for medical diagnosis, medical simulation or medical data mining; ICT specially adapted for detecting, monitoring or modelling epidemics or pandemics
- G16H50/20—ICT specially adapted for medical diagnosis, medical simulation or medical data mining; ICT specially adapted for detecting, monitoring or modelling epidemics or pandemics for computer-aided diagnosis, e.g. based on medical expert systems
-
- G—PHYSICS
- G16—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR SPECIFIC APPLICATION FIELDS
- G16H—HEALTHCARE INFORMATICS, i.e. INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR THE HANDLING OR PROCESSING OF MEDICAL OR HEALTHCARE DATA
- G16H10/00—ICT specially adapted for the handling or processing of patient-related medical or healthcare data
- G16H10/60—ICT specially adapted for the handling or processing of patient-related medical or healthcare data for patient-specific data, e.g. for electronic patient records
-
- G—PHYSICS
- G16—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR SPECIFIC APPLICATION FIELDS
- G16H—HEALTHCARE INFORMATICS, i.e. INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR THE HANDLING OR PROCESSING OF MEDICAL OR HEALTHCARE DATA
- G16H50/00—ICT specially adapted for medical diagnosis, medical simulation or medical data mining; ICT specially adapted for detecting, monitoring or modelling epidemics or pandemics
- G16H50/70—ICT specially adapted for medical diagnosis, medical simulation or medical data mining; ICT specially adapted for detecting, monitoring or modelling epidemics or pandemics for mining of medical data, e.g. analysing previous cases of other patients
-
- Y—GENERAL TAGGING OF NEW TECHNOLOGICAL DEVELOPMENTS; GENERAL TAGGING OF CROSS-SECTIONAL TECHNOLOGIES SPANNING OVER SEVERAL SECTIONS OF THE IPC; TECHNICAL SUBJECTS COVERED BY FORMER USPC CROSS-REFERENCE ART COLLECTIONS [XRACs] AND DIGESTS
- Y02—TECHNOLOGIES OR APPLICATIONS FOR MITIGATION OR ADAPTATION AGAINST CLIMATE CHANGE
- Y02A—TECHNOLOGIES FOR ADAPTATION TO CLIMATE CHANGE
- Y02A90/00—Technologies having an indirect contribution to adaptation to climate change
- Y02A90/10—Information and communication technologies [ICT] supporting adaptation to climate change, e.g. for weather forecasting or climate simulation
Abstract
The invention discloses a method for establishing a coronary artery bypass grafting perioperative risk prediction model, which combines collection and arrangement of a large number of domestic patient hospital data, a single factor and multi-factor analysis method and adopts a logisitc regression secondary prediction method to analyze to obtain a secondary prediction model, so that the variety of risk factors examined in the modeling process is enriched, the accuracy and reliability of model prediction are effectively improved, an effective assessment prediction method is provided for the coronary artery bypass grafting risk prediction of heart failure patients in China, the operation risk can be effectively reduced, and the method has positive significance for the health development of medical health industry in China.
Description
Technical Field
The invention belongs to the field of biomedicine, and particularly relates to a method for establishing a coronary artery bypass grafting perioperative risk prediction model.
Background
Coronary heart disease surgery risk prediction is a key link for identifying high-risk patients, reducing operation mortality and improving surgery curative effect. Especially in patients with heart failure, the surgical risk is obviously improved, and an accurate preoperative risk factor prediction model is more needed. At present, the main heart surgery risk scoring models at home and abroad comprise Euro SCORE, euro SCORE II, STS, sinoSCORE and the like for heart failure patient row CABG, and researches show that Euro SCORE, euro SCORE II and Sino SCORE for Chinese people can not accurately predict the hospital mortality of heart failure patient row CABG, and the mortality is obviously overestimated. The main disadvantages of the existing predictive model are: (1) The data for the creation of these models is relatively long-term (most of the 10 years old), and advances in surgical techniques and improvements in perioperative treatment levels have now significantly reduced post-CABG mortality. (2) Most models are built based on data of European populations, the population specificity is not strong, for example EuroSCORE II, the coverage population is mainly European populations, the incidence characteristics of coronary heart disease and heart failure are different from those of Chinese populations, and therefore the death rate of the Chinese heart failure line CABG operation is obviously overestimated. (3) Almost all existing models are directed to the population with common coronary heart disease, and not heart failure, such as SinoSCORE, as a predictive model established based on national data, which does not further distinguish between the population with EF <50% heart failure, so the prediction of mortality of this population is not accurate enough. Therefore, a prediction model suitable for the characteristics of the crowd in China is developed and used for predicting the peri-operative risk of the coronary artery bypass grafting of heart failure patients, and the prediction model is used as a reference and a basis for selecting a treatment method means and has important reference significance for improving the treatment success rate and reducing the operation risk.
Disclosure of Invention
The invention aims to provide a method for establishing a coronary artery bypass grafting perioperative risk prediction model, which can more accurately predict the coronary artery bypass grafting operation risk of heart failure patients, is more suitable for Chinese people, and is suitable for popularization in clinical practice.
In order to achieve the above purpose, the present invention adopts the following technical scheme:
a method for establishing a coronary artery bypass grafting perioperative risk prediction model is characterized in that the prediction model obtains relevant risk factors through regression analysis and selection of the coronary artery bypass grafting perioperative risk factors, obtains final risk factors through multi-factor regression analysis and screening of the relevant risk factors, obtains an initial prediction model through regression analysis, and finally optimizes the initial prediction model through secondary prediction to obtain the prediction model.
Further, the building of the model comprises the following steps:
(1) Collecting clinical data of patients with preoperative heart failure subjected to coronary artery bypass grafting operation and establishing a disease database;
(2) Respectively carrying out correlation analysis on the initial risk factors, selecting the correlated risk factors as research objects, wherein the selection basis is that the p value is less than 0.1 and the occurrence rate is more than 5 percent;
(3) Carrying out logistic regression analysis on the correlation risk factors in the step (2), carrying out multi-factor collinearity judgment, and removing the collinearity variables;
(4) Removing co-linear variablesThen, obtaining final risk factors, performing model fitting by using logistic regression, and obtaining regression equation partial regression coefficient beta i Constant beta in regression equation 0 Obtaining an initial prediction model;
(5) Setting a primary judgment value, performing primary prediction analysis, and dividing sample data into three groups: death group, security group, and uncertainty group;
(6) Setting a secondary judgment value, and carrying out logistic analysis on the uncertain group in the step (5) to obtain a secondary prediction model. The improvement of the initial prediction model can be realized, and the accuracy and reliability of prediction are improved.
Further, the initial risk factors of step (2) include: sex, hyperlipidemia, brain natriuretic peptide, history of thyroid dysfunction, hemoglobin, alanine aminotransferase, hypertension, body mass index, myocardial infarction, diabetes, stent implantation of cardiac vessels, elevated creatinine, cardiac surgery, history of smoking, peripheral arterial lesions, cerebrovascular events, pre-operative critical conditions, CCS grade 4, preoperative atrial fibrillation or flutter, NYHA class III or IV, left ventricular ejection fraction (35% < LVEF < 45%, LVEF < 35%), combined valve surgery, combined aortic surgery, non-optional surgery and extracorporeal circulation surgery.
Further, the related risk factors of step (2) include: sex, hyperlipidemia, brain natriuretic peptide, hemoglobin, elevated alanine Aminotransferase (ALT), body mass index, history of myocardial infarction, diabetes, elevated creatinine, history of prior cardiac surgery, cerebrovascular events, thyroid dysfunction, preoperative critical state, NYHA grade III or IV cardiac function, left ventricular ejection fraction (35% < LVEF < 45%, LVEF < 35%), concomitant valve surgery and concomitant aortic surgery.
Further, the final risk factor in step (4) includes: sex, elevated alanine Aminotransferase (ALT), brain natriuretic peptide, history of thyroid dysfunction, past heart surgery, elevated creatinine values, grade III or IV heart function, left ventricular ejection fraction, concomitant valve surgery, concomitant aortic surgery.
Further, the perioperative mortality prediction equation corresponding to the initial prediction model in the step (4) is:
further, xi in the prediction equation is a risk factor after screening, when the risk factor appears, xi=1, and when the risk factor does not appear, xi=0.
Further, the value of beta 0 in the prediction equation is-3.273.
Further, the primary judgment values in the step (5) are 0.7 and 0.3, when the predicted probability value reaches or exceeds 0.7, the judgment is made as a death group, when the predicted probability value is lower than 0.3, the judgment is made as a survival group, and the rest is an uncertain group.
Further, the secondary determination value in the step (6) is 0.6.
Compared with the existing other models, the model building process surveys and gathers the cardinal number of the diseased crowd, the database samples are more abundant, the inspected risk factors are more, the secondary prediction analysis is carried out on the uncertain crowd through the multi-step screening of the initial risk factors, the correlation risk factors and the final risk factors and the setting of the primary judgment value and the secondary judgment value, and the accuracy of model prediction is greatly improved through the multi-step screening and the secondary prediction method, so that the model is more suitable for Chinese crowd compared with the existing other prediction models, the prediction accuracy is higher, and the model has important risk guiding significance for clinical practice.
According to the invention, through expansion of a database of diseased people and a single-factor and multi-factor regression analysis method, the screened risk factors of brain natriuretic peptide, thyroid function and alanine aminotransferase are proved by experiments, and the accuracy of model prediction is improved effectively by the risk factors, so that the model prediction accuracy is stronger, and the method is more suitable for popularization in clinical practice.
According to the invention, by introducing new risk factors and combining a modeling method of secondary prediction, the prediction accuracy and reliability of the model are effectively improved, and the method has important guiding significance for clinical operation.
The risk factors brain natriuretic peptide, thyroid function and alanine aminotransferase provide more references and evaluation basis for predicting the peri-operative risk of the coronary artery bypass grafting of heart failure patients, and provide more basis and research methods for further deeply researching the reasons and mechanisms of the peri-operative risk of the coronary artery bypass grafting of heart failure patients.
Drawings
FIG. 1 is a schematic flow chart of a method for establishing a model for predicting the peri-operative risk of a coronary artery bypass grafting.
Detailed Description
The invention will be further described with reference to the following examples in conjunction with the flow diagram (fig. 1).
Example 1
The medical treatment and collection of 3659 domestic patients with complete clinical data subjected to coronary artery bypass grafting due to heart failure as modeling study objects in 2010 to 2019, and the medical treatment and collection of the integrity of the patient data and the actual collection result comprise the following steps: sex, hyperlipidemia, brain natriuretic peptide, history of thyroid dysfunction, hemoglobin, alanine aminotransferase, hypertension, body mass index, myocardial infarction, diabetes, stent implantation of cardiac vessels, elevated creatinine, cardiac surgery, smoking history, peripheral arterial lesions, cerebrovascular events, pre-operative critical state, CCS4 grade, preoperative atrial fibrillation or flutter, NYHA (heart failure) cardiac function III or IV grade, left ventricular ejection fraction (35% < LVEF < 45%, LVEF < 35%), combined valve surgery, combined aortic surgery, non-selective surgery and extracorporeal circulation surgery, 25 initial risk factors as study objects, hyperlipidemia judgment criteria, when the following fasting plasma examination index is not less than 1 item, can diagnose dyslipidemia, total Cholesterol (TC) not less than 6.2mmol/L, low density lipoprotein cholesterol (LDL-C) not less than 4.1mmol/L, triglyceride (TG) not less than 2.3mmol/L, high density lipoprotein cholesterol (HDL-C) not less than 1.0mmol/L; the brain natriuretic peptide classification criteria are: less than 50 years old, brain natriuretic peptide > 450pg/ml, between 50 and 75 years old, brain natriuretic peptide > 900pg/ml, greater than 75 years old, brain natriuretic peptide greater than 1800pg/ml; thyroid function, whether there is a history of thyroid abnormalities; hemoglobin, with < 90g/L as demarcation point; alanine aminotransferase, whether there is an increase in alanine aminotransferase; hypertension, whether there is a systolic pressure > 140mmHg or a diastolic pressure > 90mmHg; whether there is a history of myocardial infarction; whether diabetes has a history of sugar sickness; the stent implantation of the cardiac blood vessel is carried out in the prior art, and whether the stent implantation operation of the cardiac blood vessel is carried out or not; blood creatinine, preoperative blood creatinine > 176umol/L; heart surgery, whether there is heart surgery with open pericardium; a history of smoking, whether there is a history of smoking; peripheral arterial lesions, whether peripheral arterial lesions exist in the past; cerebrovascular events, with or without coma or central nervous system abnormalities for more than 24 hours, for more than 72 hours; a pre-operative critical state, whether any of ventricular tachycardia or ventricular fibrillation or sudden death from rescue; CCS grade 4, CCS angina classification grade 4; preoperative atrial fibrillation or atrial augmentation, and preoperative atrial fibrillation or atrial augmentation is not present in two weeks; the left ventricular ejection fraction is divided into two cases that LVEF is more than 35% and less than 45% and LVEF is less than or equal to 35%; combining valve operations, whether there are any valve operations combined; and merging aortic operations, and whether any merging aortic operations exist. The patient information is arranged, a patient information database is established, 3659 patients are divided into two groups according to age, sex and body quality index indexes, one group serves as a building module, the other group serves as a verification group, the number of the groups of the modeling group is 2365, the number of the persons in the verification group is 1294, the distribution conditions of the age, sex and body quality index indexes of the building module and the verification group are basically consistent, the actual mortality of the modeling group is 1.47%, and the actual mortality of the verification group is 1.42%.
Example 2
For the modeling group in example 1, single factor analysis is performed, the relationship between each single factor in 25 initial risk factors and perioperative mortality is analyzed, risk factors with p value less than 0.1 and occurrence rate more than 5% are screened, and the relevant risk factors obtained by the condition screening are as follows: sex, hyperlipidemia, brain natriuretic peptide, hemoglobin, alanine Aminotransferase (ALT) elevation, body mass index, myocardial infarction history, diabetes, elevated creatinine, past cardiac surgery, cerebrovascular events, history of thyroid dysfunction, preoperative critical state, NYHA class III or IV, left ventricular ejection fraction (35% < LVEF < 45%, LVEF < 35%), combined valve surgery and combined aortic surgery 17 risk factors were performed using SPSS 20.0.
Example 3
And (3) carrying out multi-factor collinearity judgment on 17 relevant risk factors in the step (2) by adopting logistic regression analysis, removing the collinearity variable, and removing 7 risk factors including hyperlipidemia, hemoglobin, body quality index, myocardial infarction medical history, diabetes, cerebrovascular events and preoperative critical states to obtain the complex composition comprising: gender, elevated alanine Aminotransferase (ALT), brain natriuretic peptide, history of thyroid dysfunction, past cardiac surgery, elevated creatinine levels, grade III or IV cardiac function, left ventricular ejection fraction, combined valve surgery, combined aortic surgery, 10 final risk factors (11 independent factors), 10 final risk factors and weights are shown in table 1.
TABLE 1 Risk factor regression coefficient
The method comprises the following steps: gender, elevated alanine Aminotransferase (ALT), brain natriuretic peptide, history of thyroid dysfunction, past cardiac surgery, elevated creatinine values, grade III or IV cardiac function, left ventricular ejection fraction, hemoglobin, combined valve surgery, combined aortic surgery, and performing logistic regression of 10 final risk factors to obtain partial regression coefficients βi and β0, and obtaining an initial prediction model, wherein the perioperative mortality prediction equation is:the parameters required in the equation are shown in table 1, xi is the risk factor (xi=1 if present), βi is the β coefficient of the corresponding variable in table 1, β0= -3.273, through which mortality prediction can be performed, 1000 people are selected from the validation group, divided according to age, sex and body mass indexAnd 5 groups, wherein the actual mortality of each group is consistent, 3 deaths are ensured in each 200 groups, the actual mortality of each group is 1.50%, and model prediction is carried out on the 5 groups, so that the prediction result is as follows: 1.50.+ -. 0.29% and the statistical analysis used was done with SPSS 20.0.
Example 4
In order to further improve the accuracy of model prediction, a secondary prediction model is built, on the basis of the embodiment 3, a primary judgment value is determined according to the regression prediction result of the embodiment 3, primary prediction analysis is performed, the model is built to select 0.7 and 0.3 as primary judgment values, when the prediction probability value reaches or exceeds 0.7, the model is judged to be a death group, when the prediction probability value is lower than 0.3, the model is judged to be a survival group, and the rest is an uncertain group, so that the obtained death rate prediction equation beta is obtained 1 = -5.915, mortality equation:then, carrying out secondary prediction analysis on the uncertain group, determining a new secondary judgment value according to the previous probability information and loss information by secondary prediction, judging the new judgment value of the model as a death group when the prediction probability value reaches or exceeds 0.6, judging the death group as a survival group when the prediction probability value is lower than 0.6, and obtaining a death rate prediction equation beta 2 = -4.173, quadratic predictive model mortality equation: />The secondary prediction analysis further improves the accuracy and reliability of prediction. Relevant regression analysis parameters in the predictive model are shown in Table 2.
TABLE 2 multifactor regression coefficients
Example 5
2000 people are selected from the modeling group, 5 groups are divided according to age, sex and body quality index indexes, the actual mortality of each group is guaranteed to be consistent, 6 deaths exist in each 400 people group, the actual mortality of each group is 1.50%, a secondary prediction model is adopted for model prediction, and the prediction result is that: mortality is (1.50+/-0.18)%, 1000 persons are selected from the verification group, 5 groups are divided according to age, sex and body quality index indexes, the actual mortality of each group is consistent, 3 deaths are ensured in each 200 person group, the actual mortality of each group is 1.50%, model prediction is carried out by adopting a secondary prediction model, and the prediction result is that: the mortality rate is (1.50+/-0.16)%, and the reliability and stability of the secondary prediction model are further verified by the grouping prediction results of the building module and the verification group.
The 5 groups of the above verification groups were predicted using EuroSCORE, euroSCORE II and SinoSCORE to obtain predicted mortality rates of (3.97.+ -. 0.45)%, (2.38.+ -. 0.52)% and (7.82.+ -. 0.36)%, respectively.
Through the comparison, the EuroSCORE, euroSCORE II and SinoSCORE predictive model established on the basis of European and American species data has larger deviation between the predicted result and the actual death rate of Chinese people.
Example 6
Verification experiment of the effect of three risk factors on secondary predictive model for alanine aminotransferase elevation, brain natriuretic peptide, and history of thyroid dysfunction: the verification object is 5 groupings of the verification group in example 5, with the purpose of verifying the accuracy of the secondary predictive model predictions when incorporating different risk factors. 1) 7 risk factors: sex, past cardiac surgery, elevated blood creatinine values, grade III or IV cardiac function, left ventricular ejection fraction, combined valve surgery and combined aortic surgery, plus elevated risk factor alanine aminotransferase, for a total of 8 risk factors, the predicted outcome is: (1.50±0.39)%; 2) 7 risk factors: gender, past cardiac surgery, elevated blood creatinine values, grade III or IV cardiac function, left ventricular ejection fraction, combined valve surgery and combined aortic surgery, plus brain natriuretic peptide, when total 8 risk factors, the predicted outcome is: (1.50±0.41)%; 3) 7 risk factors: sex, past cardiac surgery, elevated blood creatinine values, grade III or IV cardiac function, left ventricular ejection fraction, combined valve surgery and combined aortic surgery, plus risk factors for history of thyroid dysfunction, for a total of 8 risk factors, the predicted outcome is: (1.50±0.37)%; 4) 7 risk factors: sex, past cardiac surgery, elevated blood creatinine values, grade III or IV cardiac function, left ventricular ejection fraction, combined valve surgery and combined aortic surgery, plus two risk factors for elevated brain natriuretic peptide and alanine aminotransferase, total 9 risk factors, predicted results are: (1.50±0.33)%; 5) 7 risk factors: sex, past cardiac surgery, elevated blood creatinine values, grade III or IV cardiac function, left ventricular ejection fraction, combined valve surgery and combined aortic surgery, plus two risk factors of brain natriuretic peptide and history of thyroid dysfunction, when 9 risk factors are summed, the predicted outcome is: (1.50±0.37)%; 6) 7 risk factors: sex, past cardiac surgery, elevated blood creatinine values, grade III or IV cardiac function, left ventricular ejection fraction, combined valve surgery and combined aortic surgery, plus two risk factors, alanine aminotransferase elevation and a history of thyroid dysfunction, together with 9 risk factors, the predicted outcome is: (1.50±0.40)%; 7) Only 7 risk factors were included: the predicted results for gender, past cardiac surgery, elevated blood creatinine values, grade III or IV cardiac function, left ventricular ejection fraction, combined valve surgery and combined aortic surgery are: (1.50±0.48)%; 8) 7 risk factors: sex, past cardiac surgery, elevated blood creatinine values, grade III or IV cardiac function, left ventricular ejection fraction, combined valve surgery and combined aortic surgery, plus 3 risk factors for brain natriuretic peptide, elevated alanine aminotransferase and history of thyroid dysfunction, namely 10 final risk factors according to the invention, the model death prediction results are: (1.50.+ -. 0.16)%.
Through the analysis experiments of different combinations of the three risk factors, three risk factors can be obtained: the lack of either or both risk factors has a direct impact on the accuracy of the model's predictions, including elevated alanine aminotransferase, brain natriuretic peptide, and history of thyroid dysfunction.
When three risk factors including alanine aminotransferase elevation, brain natriuretic peptide and thyroid gland dysfunction history are added simultaneously in the 7 risk factors, the model prediction effect obtained by the establishment is best, the death prediction result is (1.50+/-0.16)%, the confidence interval is narrowest, the accuracy is highest, and the three risk factors are also further explained, the alanine aminotransferase elevation, brain natriuretic peptide and thyroid gland dysfunction history are indispensable for the establishment of a prediction model, and have important significance for the accuracy of model prediction established by the invention.
The foregoing examples are provided to facilitate understanding of the method and core ideas of the present application, and are not to be construed as limiting the present application where variations are apparent to those of ordinary skill in the art in light of the concepts of the present application.
Claims (4)
1. The method for establishing the peri-operative risk prediction model of the coronary artery bypass grafting is characterized in that the peri-operative risk prediction model obtains relevant risk factors through regression analysis and selection of the peri-operative risk factors of the coronary artery bypass grafting, obtains final risk factors through multi-factor regression analysis and screening of the relevant risk factors, obtains an initial prediction model through regression analysis, and finally optimizes the initial prediction model through secondary prediction to obtain the peri-operative risk prediction model;
the establishment of the perioperative risk prediction model comprises the following steps:
(1) Collecting clinical data of patients with preoperative heart failure subjected to coronary artery bypass grafting operation and establishing a disease database;
(2) Respectively carrying out correlation analysis on the initial risk factors, selecting the correlated risk factors as research objects, wherein the selection basis is that the p value is less than 0.1 and the occurrence rate is more than 5 percent;
(3) Carrying out logistic regression analysis on the related risk factors in the step (2), carrying out multi-factor collinearity judgment, and removing the collinearity variable;
(4) Removing the collinearity variable to obtain a final risk factor, and performing model fitting by using logistic regression to obtain a regression equation partial regression coefficient beta i Constant beta in regression equation 0 Obtaining an initial prediction model;
(5) Setting a primary judgment value, performing primary prediction analysis, and dividing sample data into three groups: death group, security group, and uncertainty group;
(6) Setting a secondary judgment value, and carrying out logistic analysis on the uncertain group in the step (5) to obtain the perioperative risk prediction model;
the initial risk factors of step (2) include: sex, hyperlipidemia, brain natriuretic peptide, history of thyroid dysfunction, hemoglobin, alanine aminotransferase, hypertension, body mass index, myocardial infarction, diabetes, stent implantation of cardiac vessels, elevated blood creatinine, cardiac surgery, history of smoking, peripheral arterial lesions, cerebrovascular events, preoperative critical conditions, CCS grade 4, preoperative atrial fibrillation or flutter, NYHA cardiac function grade III or IV, left ventricular ejection fraction, combined valve surgery, combined aortic surgery, non-selective surgery and extracorporeal circulation surgery;
the related risk factors of step (2) include: sex, hyperlipidemia, brain natriuretic peptide, hemoglobin, elevated alanine aminotransferase, body mass index, history of myocardial infarction, diabetes, elevated creatinine, past heart surgery, cerebrovascular events, history of thyroid dysfunction, preoperative critical state, NYHA heart function grade III or IV, left ventricular ejection fraction, combined valve surgery and combined aortic surgery;
the final risk factors in step (4) include: sex, elevated alanine aminotransferase, brain natriuretic peptide, history of thyroid dysfunction, past heart surgery, elevated creatinine values, NYHA heart function grade III or IV, left ventricular ejection fraction, concomitant valve surgery, concomitant aortic surgery;
and when three risk factors including alanine aminotransferase elevation, brain natriuretic peptide and thyroid dysfunction history are simultaneously added into the risk factors of the final risk factors, establishing a death prediction result of the perioperative risk prediction model to be (1.50+/-0.16)%.
2. The method for building a model for predicting peri-operative risk of coronary artery bypass grafting according to claim 1, wherein the initial model for predicting peri-operative mortality in step (4) has the following equation:
x in the perioperative mortality prediction equation i For the risk factors after screening, X is used when the risk factors appear i =1, risk factor is not present, X i =0; beta in the perioperative mortality prediction equation 0 Is-3.273, and beta in the perioperative mortality prediction equation i Is a partial regression coefficient.
3. The method for building a model for predicting the peri-operative risk of coronary artery bypass grafting according to claim 1, wherein the primary judgment values in the step (5) are 0.7 and 0.3, and when the predicted probability value is equal to or greater than 0.7, the predicted probability value is judged as a death group; when the predicted probability value is smaller than 0.3, judging the predicted probability value as a survival group; and when the predicted probability value is more than or equal to 0.3 and less than 0.7, judging the predicted probability value as an uncertain group.
4. The method for building a model for predicting peri-operative risk of coronary artery bypass grafting according to claim 1, wherein the secondary determination value in the step (6) is 0.6.
Priority Applications (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN202110077755.1A CN112820397B (en) | 2021-01-20 | 2021-01-20 | Method for establishing peri-operative risk prediction model of coronary artery bypass grafting |
Applications Claiming Priority (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN202110077755.1A CN112820397B (en) | 2021-01-20 | 2021-01-20 | Method for establishing peri-operative risk prediction model of coronary artery bypass grafting |
Publications (2)
Publication Number | Publication Date |
---|---|
CN112820397A CN112820397A (en) | 2021-05-18 |
CN112820397B true CN112820397B (en) | 2024-04-05 |
Family
ID=75858756
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
CN202110077755.1A Active CN112820397B (en) | 2021-01-20 | 2021-01-20 | Method for establishing peri-operative risk prediction model of coronary artery bypass grafting |
Country Status (1)
Country | Link |
---|---|
CN (1) | CN112820397B (en) |
Families Citing this family (2)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN113327679A (en) * | 2021-05-27 | 2021-08-31 | 上海市闵行区中心医院 | Pulmonary embolism clinical risk and prognosis scoring method and system |
CN115910365B (en) * | 2022-11-16 | 2024-02-13 | 华中科技大学同济医学院附属同济医院 | Construction method and prediction system of Eos-CRSwNP preoperative refractory evaluation model |
Citations (1)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN112837819A (en) * | 2021-01-20 | 2021-05-25 | 尹丽君 | Method for establishing acute kidney injury prediction model after coronary artery bypass grafting |
Family Cites Families (1)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
US8796948B2 (en) * | 2009-11-10 | 2014-08-05 | Lumenetix, Inc. | Lamp color matching and control systems and methods |
-
2021
- 2021-01-20 CN CN202110077755.1A patent/CN112820397B/en active Active
Patent Citations (1)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN112837819A (en) * | 2021-01-20 | 2021-05-25 | 尹丽君 | Method for establishing acute kidney injury prediction model after coronary artery bypass grafting |
Non-Patent Citations (2)
Title |
---|
心力衰竭患者冠状动脉旁路移植术围术期死亡风险评分系统的建立与对比研究;林宏远等;中国循环杂志(第5期);第444-448页 * |
病例对照研究中两阶段抽样设计与参数估计方法;沈其君等;南京铁道医学院学报(第1期);第54-57页 * |
Also Published As
Publication number | Publication date |
---|---|
CN112820397A (en) | 2021-05-18 |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
CN112837819B (en) | Method for establishing acute kidney injury prediction model after coronary artery bypass grafting operation | |
CN112820397B (en) | Method for establishing peri-operative risk prediction model of coronary artery bypass grafting | |
Wu et al. | An artificial intelligence approach to early predict non-ST-elevation myocardial infarction patients with chest pain | |
Semeraro et al. | Predicting the risk of diabetic retinopathy in type 2 diabetic patients | |
CN109378072A (en) | A kind of abnormal fasting blood sugar method for early warning based on integrated study Fusion Model | |
CN109308545A (en) | The method, apparatus, computer equipment and storage medium of diabetes probability are suffered from prediction | |
WO2021190300A1 (en) | Method for constructing ai chronic kidney disease risk screening model, and chronic kidney disease risk screening method and system | |
CN107194138B (en) | Fasting blood glucose prediction method based on physical examination data modeling | |
CN114898873A (en) | Method and system for predicting cardiovascular disease risk of diabetes mellitus pre-stage patient | |
CN111968748A (en) | Modeling method of diabetic complication prediction model | |
CN113327679A (en) | Pulmonary embolism clinical risk and prognosis scoring method and system | |
CN115019919B (en) | Deep learning-based method for establishing function prediction model after early recovery of stroke | |
Fortin et al. | Prognostic value of oxygen kinetics during recovery from cardiopulmonary exercise testing in patients with chronic heart failure | |
Shu et al. | Clinical application of machine learning-based artificial intelligence in the diagnosis, prediction, and classification of cardiovascular diseases | |
Provenzano et al. | Precision nephrology is a non-negligible state of mind in clinical research: remember the past to face the future | |
Jackson et al. | A 20-year population study of peripartum cardiomyopathy | |
CN117079810A (en) | Cardiovascular disease unscheduled re-hospitalization risk prediction method | |
Szyguła‑Jurkiewicz et al. | One‑year survival of ambulatory patients with end‑stage heart failure: the analysis of prognostic factors | |
CN114864097A (en) | Method and device for establishing aorta dissection patient postoperative death prediction model | |
Farzi et al. | Predicting serious diabetic complications using hidden pattern detection | |
CN106570346B (en) | Physiological condition evaluation factor determination method and physiological condition evaluation factor determination system | |
CN111883248B (en) | Prediction system for childhood obesity | |
Zhou et al. | A novel laboratory-based model to predict the presence of obstructive coronary artery disease comparison to coronary artery disease consortium 1/2 score, duke clinical score and diamond-forrester score in china | |
JP7463657B2 (en) | Method for assessing the possibility of cardiac recovery in patients with heart failure | |
Choi et al. | Deriving rules for identifying diabetic among individuals with metabolic syndrome. |
Legal Events
Date | Code | Title | Description |
---|---|---|---|
PB01 | Publication | ||
PB01 | Publication | ||
SE01 | Entry into force of request for substantive examination | ||
SE01 | Entry into force of request for substantive examination | ||
TA01 | Transfer of patent application right |
Effective date of registration: 20210826 Address after: 100037 Beijing City, Xicheng District No. 167 North Applicant after: FUWAI HOSPITAL, CHINESE ACADEMY OF MEDICAL SCIENCES Address before: 6-3, 417 Hengshan street, Lushunkou District, Dalian, Liaoning, 116041 Applicant before: Yin Lijun |
|
TA01 | Transfer of patent application right | ||
GR01 | Patent grant | ||
GR01 | Patent grant |